1. Field of the Invention
Embodiments of the present invention relate, in general, to inductive sensors and more particularly to a low noise inductive sensor with coupled capacitors.
2. Relevant Background
An inductive sensor is comprised of an induction loop that, when electric current passes through it, generates a magnetic field. In electromagnetism and electronics, inductance is the property of a conductor by which a change in current in the conductor “induces” (creates) a voltage (electromotive force) in both the conductor itself (self-inductance) and any nearby conductors (mutual inductance). To add inductance to a circuit, electronic components called inductors are used, typically consisting of coils of wire to concentrate the magnetic field and so that the magnetic field is linked into the circuit. The inductance of such a loop changes according to the material inside the magnetic field and since metals are much more effective inductors (conductors) than other materials the presence of metal in the magnetic field changes the current flowing through the loop. This change can be detected by sensing circuitry.
Inductive sensors are used in science, engineering, industry and medicine to measure a wide variety of physical phenomena. For example, metal detectors at airports are essentially inductive sensors as are control systems for traffic lights. Consider the pavement sensors near an intersection. These sensors can include one or more wire loops positioned within the pavement such that when a large metal mass approaches the loop, such as an automobile, the current within the loop changes. The inductive sensor described broadly above is referred to as an eddy current (EC) sensor. Eddy currents are created when a conductor experiences changes in the magnetic field. If either the conductor is moving through a steady magnetic field, or the magnetic field is changing around a stationary conductor, eddy currents will occur in the conductor. Therefore, eddy currents will be generated wherever a conducting object experiences a change in the intensity or direction of the magnetic field at any point within it, and not just at the boundaries. Eddy current sensors induce and detect eddy currents in electrically conductive bodies such as metals, liquids, industrial coatings and biological tissue. Such sensors can also measure material characteristics such as corrosion, cracking, wear and damage and are extensively used in the field of non-destructive testing (NDT).
Beyond basic metal detectors, EC sensors are particularly well suited to measuring the position or displacement of a metallic object because they are highly sensitive to position and motion while being highly insensitive to environmental factors such as dirt, water, temperature and radiation. EC sensors are often used in precision motion control systems such as the fine steering mirrors in satellite optics. In these systems the EC sensor provides motion feedback on the tip and tilt of a servo-controlled mirror used, for example, to image objects on the ground or to point a laser beam at another satellite. However, noise, or unwanted random signals, in the EC sensors causes the fine steering mirror to move unpredictably, generating “optical jitter” and degrading image quality or pointing accuracy. Therefore, there is a continued need for EC sensors with lower noise and higher accuracy.
FIG. 1 shows the arrangement of a differential eddy current sensor as would be known in the prior art. In this prior art example, the driver/demodulator block 110 drives the sensor coils, Lsns+ 120, and Lsns− 130, through the series elements Zs 140, which can be a resistor, inductor or capacitor or combination thereof. The drive signal 150, Vdrv, drives the two sensors 120, 130 with a sine wave, square wave, or similar periodic waveform generated from an oscillator 155 that is typically 500 kHz. Parallel capacitors Cp 160 resonate with the inductance of the sensors, improving linearity, temperature coefficient, and noise while reducing power consumption. The sensors are energized with an AC current at the drive signal frequency, and the magnetic field generated by the current induces eddy currents in the conductive target 170. The eddy currents generate an opposing magnetic field that can be sensed as an impedance change at the terminals of the sensor coil. The impedance change depends on the “lift-off,”175 or distance from the sensor coil to the target, so the voltages at RF+ 135 and RF− 145 vary in amplitude and phase in response to target movement. The demodulator 110 converts the AC signal at RF+ 135 and RF− 145 to DC (or baseband) at IF=V(IF+180−IF−185). IF 180, 185 may be filtered, offset, scaled or digitized in a downstream signal processing stage (not shown.)
When a target 170 is positioned precisely equidistant from the sensors 120, 130 (the null point), the sensor voltages RF+ 135 and RF− 145 are balanced with equal amplitude and equal phase. As the target 170 moves, it changes the inductance of the two sensor coils 120, 130 in a complementary way, unbalancing the voltages RF+ 135 and RF− 145. The demodulator 110 typically processes the difference (RF+−RF−) to produce a proportional DC signal at IF.
FIG. 2 shows a prior art demodulator circuit, which is also known as a standard Gilbert mixer. The demodulator circuit of FIG. 2 receives the voltages on the sensors at RF+135 and RF−145. The transistors QdiffP 270 and QdiffN 280 form a differential transconductor, converting the voltage difference (RF+−RF−) to a sinusoidal current which is fed into the switching quad comprised of Qsw1p, 210 Qsw1n, 220 Qsw2p, 230 and Qsw2n 240. These transistors are switched by the switch drive signals LO+ and LO− at the same frequency as the sensor drive signals, typically 500 kHz. The switching mixes the 500 kHz sensor signal (RF+−RF−) 135, 145 down to DC at IF+ 180 and IF− 185. Low pass filters, not shown, remove the 500 kHz signal and its harmonics producing IF=V(IF+−IF−).
The prior art demodulator shown in FIG. 2 has several shortcomings when used in an eddy current sensor system such as that shown in FIG. 1. For example, the transistors in the transconductor generate a significant amount of noise through their base resistance, base shot noise and collector shot noise. Moreover the transconductor has a limited common mode voltage range such that the circuit must be biased with high enough voltage to accommodate the full voltage swing on the sensors that appears largely in-phase at RF+ and RF−. This causes an increase in the minimum power supply voltage and therefore an increase in power dissipation.
The transconductor also adds an additional level of transistors to the switch transistors, requiring an increase in power supply voltage and power consumption, and the differential pair must be degenerated with the emitter resistors Rep 250 and Ren 255 to maintain adequate linearity and to control the temperature stability. Rep 250 and Ren 255 generate significant noise and can cause further voltage drops that require an increase in supply voltage and power consumption. While Rep 250 and Ren 255 could be replaced with inductors, inductors are usually physically larger and less temperature stable than resistors. Because Rep 250 and Ren 255 have relatively low impedance the transconductor must be biased with relatively high current to maintain linearity and temperature stability, further increasing noise and power dissipation.
Finally the finite current gain of the differential pair, QdiffP 270 and QdiffN 280, causes a loss of signal that is dependent on temperature and radiation exposure, degrading accuracy and stability, particularly in high radiation environments such as space. As can be seen above, there is a continued need for EC sensors with lower noise and higher accuracy. These and other deficiencies of the prior art are addressed by one or more embodiments of the present invention.